pith. sign in

arxiv: 2604.21800 · v1 · submitted 2026-04-23 · 🪐 quant-ph · cs.IT· math-ph· math.IT· math.MP

Variance Geometry of Exact Pauli-Detecting Codes: Continuous Landscapes Beyond Stabilizers

Arunaday Gupta , Baisong Sun , Xi He , Bei Zeng This is my paper

Pith reviewed 2026-05-09 22:28 UTC · model grok-4.3

classification 🪐 quant-ph cs.ITmath-phmath.ITmath.MP
keywords quantum error detectionPauli errorshigher-rank numerical rangesstabilizer codesnonadditive codesvariance geometryexact quantum codesKnill-Laflamme conditions
0
0 comments X

The pith

Exact Pauli-detecting codes form connected continuous families of attainable variance values rather than isolated discrete solutions.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper treats exact quantum codes that detect a fixed set of Pauli errors as objects whose existence is governed by the joint higher-rank numerical ranges of those Pauli operators. It introduces a single scalar λ* equal to the Euclidean norm of the vector of Pauli expectation values on the maximally mixed code state, and shows that this scalar organizes the space of all valid codes into connected intervals. In every unrestricted case and every symmetry-compatible case examined, the attainable λ* values fill a single closed interval whenever the set is nonempty. Stabilizer codes appear only as discrete, measure-zero points inside these intervals. The analysis combines algebraic geometry of operator compressions with numerical optimization over Stiefel manifolds and symmetry-sector decompositions.

Core claim

Exact detection of a prescribed Pauli error set is equivalent to the code projector lying in the joint higher-rank numerical range of those Pauli operators. The attainable values of the associated scalar λ* always form a single closed interval in every case analyzed, whether or not symmetry constraints are imposed; stabilizer codes occupy only isolated points within the interval. This picture is obtained by reducing the Knill-Laflamme conditions to eigenvalue interlacing relations on the compressed Pauli matrices and confirming the interval structure through exhaustive small-system enumeration and Stiefel-manifold optimization for larger systems.

What carries the argument

The scalar λ* defined as the Euclidean norm of the signature vector of Pauli expectation values on the maximally mixed code state, which serves as a one-parameter summary of the code's joint Pauli variance profile.

If this is right

  • Stabilizer codes become sparse special cases inside larger continuous families of nonadditive exact codes.
  • Code design reduces to selecting any point inside an interval of allowed variance values rather than searching discrete algebraic solutions.
  • Symmetry restrictions on the error model or code space preserve the connected-interval structure whenever the symmetry is compatible with the detection condition.
  • Numerical mapping of the full λ* spectrum becomes feasible for moderate system sizes via manifold optimization.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Code-search algorithms could be redesigned to sample continuously along the interval instead of enumerating discrete candidates.
  • The observed connectedness may extend to joint numerical ranges of other operator families beyond Pauli strings.
  • Tunable λ* could be used to optimize secondary figures of merit such as logical gate fidelity while preserving exact detection.

Load-bearing premise

The joint higher-rank numerical ranges of the Pauli operators completely determine exact detection and the chosen optimization and decomposition methods capture every attainable λ* value without hidden gaps or disconnected components.

What would settle it

An explicit list of Pauli operators and code dimension for which the attainable λ* values consist of two separated points with a nonempty open interval between them would falsify the single-interval claim.

Figures

Figures reproduced from arXiv: 2604.21800 by Arunaday Gupta, Baisong Sun, Bei Zeng, Xi He.

Figure 1
Figure 1. Figure 1: FIG. 1: Schematic behavior of the attainable spectrum Σ [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
read the original abstract

Exact quantum codes detecting a prescribed set of Pauli errors are approached through algebraic constructions--stabilizer, codeword-stabilized, permutation-invariant, topological, and related families. Geometrically, exact Pauli detection is governed by joint higher-rank numerical ranges of these Pauli operators, whose structure for rank $\geq 2$ is largely uncharted. From this viewpoint, we show that such codes often form connected continuous families rather than collections of disjoint solution regions. These families are characterized by a single scalar derived from the Knill-Laflamme conditions: denoted $\lambda^*$, it is the Euclidean norm of the signature vector of Pauli expectation values on the maximally mixed code state, and provides a one-parameter summary of the code's joint Pauli variance profile. Within these continuous landscapes, stabilizer codes occupy only discrete, measure-zero subsets of the attainable $\lambda^*$-spectrum, exposing a largely unexplored continuum of genuinely nonadditive exact codes. We establish this picture by analyzing the geometry of higher-rank operator compressions, and extend it to symmetry-restricted settings where cyclic and permutation symmetries are imposed on both the error model and the code projector. Small-system cases reveal interval, singleton, and empty regimes through eigenvalue interlacing and symmetry-sector decompositions; larger systems are treated numerically via Stiefel-manifold optimization and symmetry-adapted parameterizations. In every unrestricted and symmetry-compatible case analyzed, the attainable $\lambda^*$-spectrum forms a single closed interval whenever nonempty--although a general proof remains open. These results place stabilizer, symmetric, and nonadditive code families within a unified higher-rank variance framework, suggesting a continuous geometric perspective on the landscape of exact quantum codes.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 3 minor

Summary. The paper claims that exact Pauli-detecting quantum codes are governed by the joint higher-rank numerical ranges of the relevant Pauli operators. It introduces the scalar λ* (Euclidean norm of the Pauli-expectation signature vector on the maximally mixed code state, obtained from the Knill-Laflamme conditions) as a one-parameter summary of the code's variance profile. Small-system cases are analyzed via eigenvalue interlacing and symmetry-sector decompositions to reveal interval, singleton, and empty regimes; larger cases use Stiefel-manifold optimization with symmetry-adapted parameterizations. The central empirical result is that, in every unrestricted and symmetry-compatible case examined, the attainable λ*-spectrum forms a single closed interval whenever nonempty, although a general proof is left open. Stabilizer codes appear as discrete, measure-zero subsets within these continuous families.

Significance. If the interval property holds, the work supplies a unified geometric framework that embeds stabilizer, codeword-stabilized, permutation-invariant, and nonadditive exact codes inside continuous variance landscapes, showing that exact Pauli detection is far more abundant than discrete constructions suggest. The explicit use of interlacing arguments for small cases and symmetry decompositions provides concrete, verifiable support; the open acknowledgment of the missing general proof is appropriately cautious. These elements could guide systematic searches for new code families and shift emphasis from isolated codes to their attainable continua.

major comments (1)
  1. [Numerical treatment of larger systems (Stiefel optimization and symmetry-adapted parameterizations)] The central claim that the attainable λ*-spectrum is always a single closed interval rests on the completeness of the Stiefel-manifold optimization for larger systems. The manuscript provides no report of the number of random initializations, basin-hopping or multi-start diagnostics, convergence tolerances, or independent algebraic cross-checks that would exclude the possibility of missed disconnected components or additional intervals outside the sampled basins. Because the general proof is explicitly left open, this numerical coverage is load-bearing for the reported empirical result.
minor comments (3)
  1. [Abstract] Abstract: the phrase 'genuinely nonadditive exact codes' is used without a precise definition distinguishing it from the codeword-stabilized and permutation-invariant families already listed earlier in the same paragraph.
  2. [Introduction / definition of λ*] The definition of λ* as the Euclidean norm of the signature vector is clear, but the text could explicitly note that this scalar is a derived summary statistic rather than an independent parameter, to avoid any appearance of circularity with the Knill-Laflamme conditions.
  3. [Small-system cases] Small-system analysis: while eigenvalue interlacing is invoked, the manuscript does not state the precise exclusion criteria used to certify that a given interval is empty or a singleton; adding a short remark on the numerical tolerance or algebraic certificate would improve reproducibility.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the thoughtful summary and for identifying the need for greater transparency in the numerical evidence supporting the interval property. We address the single major comment below and will revise the manuscript to strengthen the documentation of the optimization procedures.

read point-by-point responses

  1. Referee: [Numerical treatment of larger systems (Stiefel optimization and symmetry-adapted parameterizations)] The central claim that the attainable λ*-spectrum is always a single closed interval rests on the completeness of the Stiefel-manifold optimization for larger systems. The manuscript provides no report of the number of random initializations, basin-hopping or multi-start diagnostics, convergence tolerances, or independent algebraic cross-checks that would exclude the possibility of missed disconnected components or additional intervals outside the sampled basins. Because the general proof is explicitly left open, this numerical coverage is load-bearing for the reported empirical result.

    Authors: We agree that the current description of the Stiefel-manifold optimization is insufficiently detailed given that the general proof remains open. The manuscript indeed omits explicit reporting of the number of random initializations, multi-start or basin-hopping diagnostics, convergence tolerances, and any algebraic cross-checks. In the revised version we will add a dedicated paragraph (or short appendix) describing the numerical protocol in full, including the initialization strategy, convergence criteria, and verification steps used for the symmetry-adapted parameterizations. This addition will allow readers to evaluate the coverage of the sampled basins directly. We note that the small-system cases continue to rest on exact eigenvalue-interlacing arguments rather than numerics, providing an independent anchor for the overall picture. revision: yes

Circularity Check

0 steps flagged

No significant circularity; λ* spectrum analysis is an independent geometric and numerical study of the defined quantity.

full rationale

The paper defines λ* directly from the Knill-Laflamme conditions as the Euclidean norm of the Pauli-expectation signature vector on the maximally mixed code state, then analyzes the attainable values of this quantity via higher-rank numerical ranges, eigenvalue interlacing, symmetry decompositions, and Stiefel-manifold optimization. This constitutes a study of the range of a well-defined function over the set of exact detecting codes rather than any derivation in which a result is forced by construction from its own inputs. No self-citations, fitted parameters relabeled as predictions, ansatzes smuggled via prior work, or uniqueness theorems are invoked in the provided text to support the central claims. The interval observation is presented as an empirical finding in analyzed cases with the general proof left open, leaving the derivation chain self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the standard Knill-Laflamme conditions for exact error detection and the mathematical properties of higher-rank numerical ranges; no new free parameters, invented entities, or ad-hoc axioms are introduced beyond these domain assumptions.

axioms (1)
  • domain assumption Knill-Laflamme conditions determine exact Pauli detection
    Invoked to define the signature vector whose norm yields λ*.

pith-pipeline@v0.9.0 · 5612 in / 1361 out tokens · 38693 ms · 2026-05-09T22:28:07.686355+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

60 extracted references · 60 canonical work pages · 1 internal anchor

  1. [1]

    Consider the rank-2 family of codewords |0L⟩=a|00⟩+b|01⟩, |1L⟩=X 1 |0L⟩=a|10⟩+b|11⟩, (65) wherea∈R,b∈C, and|a| 2 +|b| 2 = 1

    Continuous interval. Consider the rank-2 family of codewords |0L⟩=a|00⟩+b|01⟩, |1L⟩=X 1 |0L⟩=a|10⟩+b|11⟩, (65) wherea∈R,b∈C, and|a| 2 +|b| 2 = 1. Define P(a, b) :=|0 L⟩ ⟨0L|+|1 L⟩ ⟨1L|.(66) A direct computation gives the scalar compressions P(a, b)X 2 P(a, b) = 2ℜ(¯ab)P(a, b), P(a, b)Y 2 P(a, b) = 2ℑ(¯ab)P(a, b). (67) Consequently, for the single-observab...

    work page

  2. [2]

    At the opposite extreme, some admissible error sets force all detectable signature coordinates to vanish, yield- ing Σ2(E) ={0}

    Extremal singleton values. At the opposite extreme, some admissible error sets force all detectable signature coordinates to vanish, yield- ing Σ2(E) ={0}. A canonical example is the even-parity projector Peven :=|00⟩ ⟨00|+|11⟩ ⟨11|.(69) Whenever a Pauli observableEflips the parity subspace, i.e. maps Ran(Peven) into its orthogonal complement, one hasP ev...

    work page

  3. [3]

    A fundamental constraint is eigenvalue interlacing: ifE has eigenvaluesλ 1 ≥λ 2 ≥λ 3 onH 0, then any scalar compressionP EP=αPmust satisfyα=λ 2

    Swap-basis case In the swap-basis setting Ran(P)⊆ H 0, the compres- sion of a Pauli observable is a rank-2 Hermitian compres- sion inside the three-dimensional invariant subspaceH 0. A fundamental constraint is eigenvalue interlacing: ifE has eigenvaluesλ 1 ≥λ 2 ≥λ 3 onH 0, then any scalar compressionP EP=αPmust satisfyα=λ 2. Many Pauli restrictions toH 0...

    work page

  4. [4]

    Since ev- ery swap-basis projector is in particular swap-projector- symmetric, it remains only to prove the reverse inclusion for attainable norms

    Swap-projector case We now show that, forn= 2, enforcing swap symme- try at the projector level does not enlarge the attainable signature norms beyond the swap-basis case. Since ev- ery swap-basis projector is in particular swap-projector- symmetric, it remains only to prove the reverse inclusion for attainable norms. Let |ϕ±⟩:= 1√ 2(|00⟩ ± |11⟩),(83) so ...

    work page

  5. [5]

    ThusP= ΨΨ † ranges over all rank-2 projec- tors with Ψ∈St(8,2), and detectability meansP E aP= αaPfor every listed Pauli operator

    Unrestricted random tuples We first impose no symmetry constraint on the code projector. ThusP= ΨΨ † ranges over all rank-2 projec- tors with Ψ∈St(8,2), and detectability meansP E aP= αaPfor every listed Pauli operator. We sampled 1002 random tuples: 334 withm= 6, 334 withm= 7, and 334 withm= 8. The resulting picture is strikingly uniform. In every case t...

    work page

  6. [6]

    LetTbe the cyclic shift, T|b 1b2b3⟩=|b 3b1b2⟩,(119) and letH 0 be the +1 eigenspace ofT

    Random tuples with an external cyclic-+1restriction We next imposestate-levelcyclic symmetry on the code while keeping the Pauli tuple random and asym- metric. LetTbe the cyclic shift, T|b 1b2b3⟩=|b 3b1b2⟩,(119) and letH 0 be the +1 eigenspace ofT. Forn= 3 this is the four-dimensional space H0 = span n |000⟩,|W⟩:= |001⟩+|010⟩+|100⟩√ 3 , |W⟩:= |011⟩+|101⟩+...

    work page 2000

  7. [7]

    Thus, already at n= 3, the difference between symmetry-compatible and externally imposed symmetry constraints is visible at the level of the scalar signature spectrum

    An exact disconnected cyclic-+1spectrum A representative disconnected example in Table II is the tuple Edisc = (Y XX, XXI, Y XZ, Y IX, IZI) (123) which has anexactlydisconnected cyclic-+1 spectrum: the only attainable values are 0 and 1. Thus, already at n= 3, the difference between symmetry-compatible and externally imposed symmetry constraints is visibl...

    work page

  8. [8]

    In each case, the endpoint values are realized by explicit rank-2 stabilizer codes detecting the listed Pauli errors

    Code projector with no symmetry Without imposing any symmetry on the code projec- tor, the four families above already realize the basic in- terval and rigidity patterns that recur throughout the paper. In each case, the endpoint values are realized by explicit rank-2 stabilizer codes detecting the listed Pauli errors. For a stabilizer S=⟨g 1, g2⟩(151) wi...

    work page

  9. [9]

    LetTbe the one-step cyclic shift and write ω=e 2πi/3

    Code projector with cyclic symmetry We now impose cyclic symmetry on these same four families. LetTbe the one-step cyclic shift and write ω=e 2πi/3. The Hilbert space decomposes as H=H 0 ⊕ Hω ⊕ Hω2 ,(184) with dimH 0 = 4 and dimH ω = dimH ω2 = 2. For the present four families, the nontrivial charge sectors do not contribute additional cyclic detecting pro...

    work page

  10. [10]

    Letn= 5 andK= 2, and denote byTthe cyclic shift, a 5-cycle acting on (C 2)⊗5

    Interval expansion in((5,2,2))codes. Letn= 5 andK= 2, and denote byTthe cyclic shift, a 5-cycle acting on (C 2)⊗5. We write P cyc for the sum over the five cyclic shifts of a computational basis string. When each logical basis vector is required to lie in the +1 eigenspace ofT, so thatT|0 L⟩=|0 L⟩andT|1 L⟩=|1 L⟩, the attainable signature norm is analytica...

    work page

  11. [11]

    ForE={X i, Yi, Zi}5 i=1 and any fixed orderingE, the cyclic-basis spectrum behaves differently from the inter- val expansion seen for ((5,2,2))

    Expansion from an isolated point in((5,3,2)). ForE={X i, Yi, Zi}5 i=1 and any fixed orderingE, the cyclic-basis spectrum behaves differently from the inter- val expansion seen for ((5,2,2)). Here, imposingstate- levelcyclic symmetry forces each logical basis vector to lie in the +1 eigenspace of the cyclic shiftT, and the at- tainable value ofλ ∗(P) colla...

    work page

  12. [12]

    For ((4,4,2)) and ((5,4,2)), the +1 cyclic sector does not appear to contain a rank-4 projector satisfying the KL constraints, so no state-level cyclic basis code is ob- tained

    State–projector existence gap in((4,4,2))and((5,4,2)). For ((4,4,2)) and ((5,4,2)), the +1 cyclic sector does not appear to contain a rank-4 projector satisfying the KL constraints, so no state-level cyclic basis code is ob- tained. Nevertheless, a cyclic-symmetric projector may still exist. For ((4,4,2)), one such exact cyclic-symmetric projector atλ ∗(P...

    work page

  13. [13]

    Asymmetric interval in((5,2))codes withE asym 5,2 . Atn= 5 andK= 2, we consider the hierarchical asymmetric family E:=E asym 5,2 ={X i, Yi, Zi}5 i=1 ∪ {ZiZj}1≤i<j≤5.(239) Unlike the distance-2 case above, this error set in- cludes all two-bodyZ-type correlators, so the KL con- ditions simultaneously constrain single-qubit and two- qubitZ-parity compressio...

    work page

  14. [14]

    For this family there is no distance parameter d; rather, the detectable set is specified directly

    A closed interval for((5,2))code withE mix 5 We next consider the symmetry-compatible detectable family E:=E mix 5 ={Xi, Yi, Zi}5 i=1 ∪ {XiXj, Z iZj, X iZj, Z iXj}1≤i<j≤5, (247) which consists of 55 distinct Hermitian Pauli observ- ables. For this family there is no distance parameter d; rather, the detectable set is specified directly. Unre- stricted Sti...

    work page

  15. [15]

    State–projector existence gap forE={X i, Yi, Zi}. For the parameter sets ((4,3,2)), ((4,4,2)), ((5,4,2)), and ((5,5,2)), state-level permutation invariance forces each basis vector into the Dicke subspace, which is too restrictive to supportKmutually orthonormal vectors satisfying the KL conditions; hence Σbasis K,Sn(E) =∅in each case. At the projector le...

    work page

  16. [16]

    1− 5 √ 7 16 , 3 8 # .(277) Substituting (277) into (275) yields λ∗2 ∈ 40−15 √ 7, 5 2 ≈[ 0.3137,2.5 ],(278) and therefore λ∗ ∈

    Asymmetric permutation-invariant codes forE asym 5,2 . For the same asymmetric familyE=E asym 5,2 studied under cyclic symmetry in Sec. VIII A 4, both state-level and projector-level permutation invariance lead to the same signature spectrum. In contrast with the cyclic case, this permutation-invariant spectrum has a strictly positive lower endpoint. A st...

    work page

  17. [17]

    with symmetry

    Permutation-invariant codes forE mix 5 . We now examine the mixed error familyE=E mix 5 of (247), whose cyclic-invariant spectrum was determined in Sec. VIII A 5, under full permutation invariance. Re- stricting the code projector to the fully permutation- invariant subspace, the problem reduces to the Dicke (spin-5/2) irreducible representation Hsym = sp...

    work page

  18. [18]

    A. R. Calderbank and P. W. Shor, Good quantum error-correcting codes exist, Physical Review A54, 1098 (1996)

    work page 1996

  19. [19]

    A. M. Steane, Error correcting codes in quantum theory, Physical Review Letters77, 793 (1996)

    work page 1996

  20. [20]

    Gottesman,Stabilizer Codes and Quantum Error Cor- rection, Ph.D

    D. Gottesman,Stabilizer Codes and Quantum Error Cor- rection, Ph.D. thesis, California Institute of Technology, Pasadena, California (1997)

    work page 1997

  21. [21]

    Cross, G

    A. Cross, G. Smith, J. A. Smolin, and B. Zeng, Code- word stabilized quantum codes, IEEE Transactions on Information Theory55, 433 (2009)

    work page 2009

  22. [22]

    Ouyang, Permutation-invariant quantum codes, Phys- ical Review A90, 062317 (2014)

    Y. Ouyang, Permutation-invariant quantum codes, Phys- ical Review A90, 062317 (2014)

    work page 2014

  23. [23]

    Ouyang and J

    Y. Ouyang and J. Fitzsimons, Permutation-invariant codes encoding more than one qubit, Physical Review A93, 042340 (2016)

    work page 2016

  24. [24]

    Dennis, A

    E. Dennis, A. Kitaev, A. Landahl, and J. Preskill, Topological quantum memory, Journal of Mathematical Physics43, 4452 (2002)

    work page 2002

  25. [25]

    Bengtsson and K

    I. Bengtsson and K. ˙Zyczkowski,Geometry of Quan- tum States: An Introduction to Quantum Entanglement (Cambridge University Press, 2017)

    work page 2017

  26. [26]

    E. M. Rains, R. H. Hardin, P. W. Shor, and N. J. A. Sloane, A nonadditive quantum code, Phys. Rev. Lett. 79, 953 (1997)

    work page 1997

  27. [27]

    M. Du, C. Zhang, Y.-T. Poon, and B. Zeng, Characteriz- ing quantum codes via the coefficients in Knill–Laflamme conditions, npj Quantum Information12, 13 (2026)

    work page 2026

  28. [28]

    Knill and R

    E. Knill and R. Laflamme, Theory of quantum error- correcting codes, Physical Review A55, 900 (1997)

    work page 1997

  29. [29]

    Li and Y.-T

    C.-K. Li and Y.-T. Poon, Generalized numerical ranges and quantum error correction, Journal of Operator The- ory66, 335 (2011)

    work page 2011

  30. [30]

    M.-D. Choi, D. W. Kribs, and K. Zyczkowski, Higher- rank numerical ranges and compression problems, Linear Algebra and its Applications418, 828 (2006)

    work page 2006

  31. [31]

    M.-D. Choi, M. Giesinger, J. A. Holbrook, and D. W. Kribs, Geometry of higher-rank numerical ranges, Linear and Multilinear Algebra56, 53 (2008)

    work page 2008

  32. [32]

    C.-K. Li, Y. T. Poon, and N.-S. Sze, Higher rank nu- merical ranges and low rank perturbations of quantum channels, Journal of Mathematical Analysis and Appli- cations348, 843 (2008)

    work page 2008

  33. [33]

    Gau, C.-K

    H.-L. Gau, C.-K. Li, Y.-T. Poon, and N.-S. Sze, Higher rank numerical ranges of normal matrices, SIAM Journal on Matrix Analysis and Applications32, 23 (2011)

    work page 2011

  34. [34]

    H. J. Woerdeman, The higher rank numerical range is convex, Linear and Multilinear Algebra56, 65 (2008)

    work page 2008

  35. [35]

    H. R. Afshin, M. A. Mehrjoofard, and A. Zare Chavoshi, Pauli group: Classification and joint higher rank numer- ical range, Linear Algebra and its Applications549, 136 (2018)

    work page 2018

  36. [36]

    Z. Xu, R. Schwonnek, and A. Winter, Bounding the joint numerical range of pauli strings by graph parameters, PRX Quantum5, 020318 (2024)

    work page 2024

  37. [37]

    Z.-P. Xu, J. Wang, Q. Ye, G. Koßmann, R. Schwonnek, and A. Winter, Simultaneous variances of pauli strings, weighted independence numbers, and a new kind of per- fection of graphs, arXiv:2511.13531 (2025)

    work page arXiv 2025

  38. [38]

    Ioffe and M

    L. Ioffe and M. M´ ezard, Asymmetric quantum error- correcting codes, Physical Review A75, 032345 (2007)

    work page 2007

  39. [39]

    Jackson, M

    T. Jackson, M. Grassl, and B. Zeng, Codeword stabilized quantum codes for asymmetric channels, in2016 IEEE International Symposium on Information Theory (ISIT) (2016) pp. 2264–2268

    work page 2016

  40. [40]

    Beigi, J

    S. Beigi, J. Chen, M. Grassl, Z. Ji, Q. Wang, and B. Zeng, Symmetries of Codeword Stabilized Quantum Codes, in 8th Conference on the Theory of Quantum Computation, Communication and Cryptography (TQC 2013), Leib- niz International Proceedings in Informatics (LIPIcs), Vol. 22, edited by S. Severini and F. Brandao (Schloss Dagstuhl – Leibniz-Zentrum f¨ ur I...

    work page 2013

  41. [41]

    Pollatsek and M

    H. Pollatsek and M. B. Ruskai, Permutationally invariant codes for quantum error correction, Linear Algebra and its Applications392, 255 (2004)

    work page 2004

  42. [42]

    Li and N.-S

    C.-K. Li and N.-S. Sze, Canonical forms, higher rank nu- merical ranges, totally isotropic subspaces, and matrix equations, Proceedings of the American Mathematical Society136, 3013 (2008)

    work page 2008

  43. [43]

    P. K. Sarvepalli, A. Klappenecker, and M. R¨ otteler, Asymmetric quantum LDPC codes, inProceedings of the 2008 IEEE International Symposium on Information Theory (ISIT 2008)(2008) pp. 305–309. 30

    work page 2008

  44. [44]

    A. S. Fletcher, P. W. Shor, and M. Z. Win, Channel- adapted quantum error correction for the amplitude damping channel, IEEE Transactions on Information Theory54, 5705 (2008)

    work page 2008

  45. [45]

    Lang and P

    R. Lang and P. W. Shor, Nonadditive quantum error cor- recting codes adapted to the amplitude damping channel, arXiv:0712.2586 (2007)

    work page arXiv 2007

  46. [46]

    P. W. Shor, G. Smith, J. Smolin, and B. Zeng, High performance single-error-correcting quantum codes for amplitude damping, IEEE Transactions on Information Theory57, 7180 (2011)

    work page 2011

  47. [47]

    M. F. Ezerman and M. Grassl, Asymmetric quantum codes detecting a single amplitude error, inProceedings of the 2013 IEEE International Symposium on Informa- tion Theory (ISIT 2013)(2013) pp. 922–926

    work page 2013

  48. [48]

    Cabello, S

    A. Cabello, S. Severini, and A. Winter, Graph-theoretic approach to quantum correlations, Phys. Rev. Lett.112, 040401 (2014)

    work page 2014

  49. [49]

    Boreland, I

    G. Boreland, I. G. Todorov, and A. Winter, Informa- tion theoretic parameters of non-commutative graphs and convex corners, Illinois Journal of Mathematics66, 123 (2022)

    work page 2022

  50. [50]

    Boreland, I

    G. Boreland, I. G. Todorov, and A. Winter, Sand- wich theorems and capacity bounds for non-commutative graphs, Journal of Combinatorial Theory, Series A177, 105302 (2021)

    work page 2021

  51. [51]

    R. Duan, S. Severini, and A. J. Winter, Zero-error communication via quantum channels, noncommutative graphs, and a quantum lov´ asz number, IEEE Transac- tions on Information Theory59, 1164 (2013)

    work page 2013

  52. [52]

    Serre,Linear Representations of Finite Groups (Springer, 1977)

    J.-P. Serre,Linear Representations of Finite Groups (Springer, 1977)

    work page 1977

  53. [53]

    Fulton and J

    W. Fulton and J. Harris,Representation Theory: A First Course, Graduate Texts in Mathematics, Vol. 129 (Springer, New York, 1991)

    work page 1991

  54. [54]

    R. H. Dicke, Coherence in spontaneous radiation pro- cesses, Physical Review93, 99 (1954)

    work page 1954

  55. [55]

    C. Cao, C. Zhang, Z. Wu, M. Grassl, and B. Zeng, Quan- tum variational learning for quantum error-correcting codes, Quantum6, 828 (2022)

    work page 2022

  56. [56]

    Absil, R

    P.-A. Absil, R. Mahony, and R. Sepulchre,Optimization Algorithms on Matrix Manifolds(Princeton University Press, Princeton, NJ, 2008)

    work page 2008

  57. [57]

    D. P. Kingma and J. Ba, Adam: A method for stochastic optimization, arXiv:1412.6980 (2015)

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  58. [58]

    numqi developers, numqi/numqi: v0.6.0 (2025)

    work page 2025

  59. [59]

    Transversal gates in nonadditive quantum codes,

    C. Zhang, Z. Wu, S. Huang, and B. Zeng, Transversal gates in nonadditive quantum codes, arXiv:2504.20847 (2025)

    work page arXiv 2025

  60. [60]

    Laflamme, C

    R. Laflamme, C. Miquel, J. P. Paz, and W. H. Zurek, Perfect quantum error correcting code, Physical Review Letters77, 198 (1996)

    work page 1996